JPH10240951A - Adaptive outline encoding method and device therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce transmission data amounts by adaptively encoding the outline of an object expressed with a video signal by using an outline movement estimation technique based on a difference between a previous outline and a present outline. SOLUTION: A code selecting part 100 outputs first and second mode selection signals MS1 and MS2 to a line L20. A selecting part 150 and a switch 170 transmits intra-encoded outline information or inter-encoded outline information according to the mode selection signals inputted through the line L20. When the first mode selection signal MS1 is inputted, the inter-encoded outline information is selected, transmitted to a transmitter for being transmitted, and also supplied through the switch 170 to an inter-decoding part 180A. On the other hand, when the second mode selection signal MS2 is inputted, the intra-encoded outline information is selected, transmitted to the transmitter for being transmitted, and supplied through the switch 170 to an intra-decoding part 180B.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for encoding a contour of an object represented by a video signal, and more particularly, to adaptively encoding a contour using a contour motion estimation technique. The present invention relates to an adaptive contour coding method and apparatus capable of reducing the amount of data to be transmitted.

[0002]

2. Description of the Related Art In a digital television system such as a videophone, an electronic conference, and a high-definition television system, a video line signal of a video frame signal is composed of a series of digital data called "pixels". It takes a lot of digital data to represent. However, due to the limited available frequency bandwidth on a normal transmission channel, to transmit large amounts of digital data over that channel,
In particular, for low bit rate video signal encoding systems such as video telephony and teleconferencing, various data compression techniques must be used to compress or reduce the amount of data to be transmitted.

In a low bit rate video signal coding system, one of the methods for coding a video signal is a so-called object-oriented analysis / synthesis coding method (Object-Oriented An).
analysis-Synthesis coding technique) (Michael
Hotter's paper, `` Object-Oriented Analysis-Synthesi
s Coding Based on Moving Two-Dimensional Object
s ", Signal Processing: Image Communication 2, 409
-428, October 1990).

According to this object-oriented analysis / synthesis coding technique, an input video signal is divided into a plurality of objects, and a parameter consisting of three sets defining the motion, the outline, and the pixel data of each object is obtained. , Through different coding channels.

When processing the contour of an object, the contour information is important for analyzing and synthesizing the shape of the object. A typical encoding method for representing contour information is a chain encoding method. Although this chain encoding method does not lose contour information, it has the disadvantage of requiring a large number of bits to represent that information.

In this connection, several contour information coding methods such as polygon approximation and B-spline approximation have been proposed. Among them, the polygon approximation has the disadvantage that the outline is coarsely expressed, and the B-spline approximation can express the outline more accurately, but requires a higher-order polynomial to reduce the approximation error. However, there is a disadvantage in that the overall calculation amount of the video signal encoder is increased.

[0007] One of the methods for avoiding the problem of coarse contour representation and an increase in the amount of calculation that occurs during the approximation process is a contour approximation method using discrete sine transform (DST).

With such a technique, it is possible to alleviate the roughness of the outline representation and the complexity of calculation and to reduce the amount of data to be transmitted to some extent.
In order for a low bit rate codec system having a transmission channel width of b / s to function well, it is necessary to further reduce the amount of data to be transmitted.

[0009]

SUMMARY OF THE INVENTION Accordingly, it is a primary object of the present invention to provide a method for estimating an object represented by a video signal using a contour motion estimation technique based on a difference between a front contour and a current contour. An object of the present invention is to provide an adaptive contour encoding method and apparatus capable of adaptively encoding a contour and further reducing the amount of data to be transmitted.

[0010]

According to an embodiment of the present invention, there is provided an image having a plurality of outlines represented by outline image data representing the position of an outline pixel. An adaptive contour encoding method for encoding a frame signal adaptively in contour units, wherein a first step of assigning index data to each of the contours in the processing order thereof, A second method for generating current contour information representing contour video data and index data
And a third step of generating front contour information representing contour video data and index data of each front contour processed before the processing of the current contour, and a step of generating front contour information and index information of each of the front contours. A fourth step of generating selection information and a first or second mode selection signal based on the front contour information for the front contour, and the current contour based on the selection information in response to the first mode selection signal. A fifth step of inter-encoding the line information, a sixth step of intra-encoding the current outline information in response to the second mode selection signal, and the inter-encoded outline information or the intra code And a seventh step of generating the converted contour information as coded contour information.

According to another embodiment of the present invention, a video frame signal having a plurality of contours represented by contour video data representing the position of a contour is adaptively encoded in contour units. Adaptive contour encoding device,
Indexing each of the contours in their processing order,
Index assigning means for assigning index data to the contour, first contour information generating means for producing contour image data representing the contour video data and index data of the current contour, and first contour information generating means for producing the current contour. Second contour information generating means for sequentially storing and generating contour image data and index data representing the contour data of each preceding contour processed before the processing, and the current contour information and all of the previous contour information. A mode selection means for generating selection information and a mode selection signal based on the previous contour information on the contour, and inter-coding the current contour information in accordance with the selection information to obtain an inter-coded contour Inter-coding means for generating line information; intra-coding means for intra-coding the current contour information to generate intra-coded contour information; Selecting means for selecting the inter-encoded contour information or the intra-encoded contour information as encoded contour information to be transmitted according to a mode selection signal. An adaptive contour coding device is provided.

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below in detail with reference to the drawings.

Referring to FIG. 1, there is shown a schematic block diagram of an adaptive contour encoding apparatus for encoding a video frame signal comprising a plurality of contours in contour units according to the present invention. . Each contour is represented by contour video data representing the position of a contour pixel forming the contour.
As shown in the figure, the video frame signal is output from the contour detection unit 16.
Input to 0.

The contour detection section 160 assigns index data to each contour included in the video frame signal in the processing order, and outputs contour video data and contour information having index data for each contour. I do. That is, according to the first control signal S1, the contour detection unit 160 sets the contour information defining each contour as the current contour information, via the line L10 to the intra encoding unit 130,
The data is sent to the mode selector 100 and the inter encoder 140.

[0015] First, the intra encoder 130 performs a chain encoding method, a polygon approximation, and a discrete sine transform (DS).
T) The current contour is intra-encoded using one of the known contour encoding methods such as the contour encoding method employing T), and the intra-encoded contour information is output on a line L50. .

On the other hand, the mode selection unit 100 outputs the first or second mode selection signal MS1 or MS2 on the line L20 in addition to the first and second control signals S1 and S2, the current contour information and all the previous contours. The selection information based on the front contour information on the line is output on the line L30. here,
The front contour is processed before the processing of the current contour and stored in the contour storage unit 120. This contour storage unit 120
Outputs the stored front contour information for each front contour on the line L70 in order according to the second control signal S2.

FIG. 2 is a detailed block diagram of the mode selection section 100. The current contour information and the previous contour information input to the mode selection unit 100 via the lines L10 and L70, respectively, are supplied to the contour length comparison unit 101 and the centroid matching unit 102, respectively.

The contour length comparing unit 101 compares the length of the current contour with the length of the previous contour, and supplies a corresponding comparison signal to the centroid matching unit 102. Here, the length of the contour is determined, for example, by the number of contour pixels forming the contour. In particular, if the difference between the length of the current contour and the length of the previous contour is within ± 10% of the length of the current contour, a comparison signal is provided, otherwise, No comparison signal is provided.

The centroid matching unit 102 includes a contour line length comparing unit 1
It operates based on the comparison signal from 01, the previous contour information and the current contour information. When the comparison signal is input, the centroid matching unit 102 assigns each centroid to the front contour line and the current contour line, for example, based on the previous contour information and the current contour line information. After averaging the position coordinates of the contour pixels, a candidate displacement representing a spatial displacement between the centroid of the current contour and the centroid of the front contour is calculated, and all contours on the front contour are calculated. The pixel is shifted by the candidate displacement to obtain a predicted contour, and the candidate displacement, predicted contour information and current contour information are supplied to the perturbation check unit 103.

The perturbation check unit 103 includes the centroid matching unit 10
Based on the candidate displacement from No. 2, the predicted contour information and the current contour information, the predicted contour is shifted in the horizontal and vertical directions by one or more pixels along the current contour to obtain an optimal contour for the predicted contour. It is determined by adding the candidate displacement to the spatial displacement between the optimal contour and the current contour by detecting the line, i.e. the contour with the smallest error value representing the difference between it and the current contour. Generate an optimal displacement and an error value for the optimal contour.

Referring to FIGS. 4A and 4B, there is shown a schematic diagram for explaining a process of detecting an error value with respect to an optimum contour according to a preferred embodiment of the present invention.

As shown in FIG. 4A, the error value of the optimal contour is defined as a non-overlapping area (ie, a hatched area surrounded by the current contour 10 and the optimal contour 30). According to a preferred embodiment of the present invention, the error value is defined as the number of pixels present in the non-overlapping area instead of the non-overlapping area.

On the other hand, referring to FIG. 4B, the error value of the optimum contour is obtained from the centroid of the current contour 10. First, the perturbation checking unit 103 draws, from the centroid of the current contour line 10, a plurality of (for example, 12) radially extending segments starting from a predetermined reference line and forming an equiangular angle with each other,
Intersections at which the K adjacent two radial lines at an angle of 2π / K intersect the current contour line 10 and the optimum contour line 30 (for example, intersections C1 to C1 on the current contour line 10)
2. Intersection points P1 to P12) on the optimal contour line 30 are detected,
The sum of absolute values of displacement between intersections on each radial line segment is calculated as an error value.

Referring again to FIG. 2, the error value of the optimal contour is supplied to the first comparator 104 as the current error value.
The index data and the optimal displacement of the optimal contour are supplied to the selector 106 as current matching information.

The first comparing section 104 includes the perturbation checking section 10
3 to determine the minimum value by comparing the error value of the optimal contour line input from step 3, ie, the current error value, with the previous error value input via the line L101, and determine this minimum value as the selected error value. And supplies the first or second selector control signal to the selector based on the result of the comparison process. For example, if the current error value is smaller than the previous error value, a first selector control signal is generated; otherwise, a second selector control signal is generated.

The selector 106 supplies the current matching information from the perturbation check unit 103 to the second memory unit 107 according to the first selector control signal, and sets the line L102 according to the second selector control signal. The pre-alignment information input via the second memory unit 107 is supplied to the second memory unit 107.

The selected error value from the first comparing unit 104 is stored in the first memory unit 105 and is output as a previous error value on a line L101. Further, the selected error value is transferred to the second comparing section 108 according to the third control signal S3 from the controller 109.

The second comparing section 108 compares the selected error value from the first memory section 105 with a predetermined threshold value TH1, and performs an error comparison only when the selected error value is equal to or smaller than the threshold value TH1. The signal is supplied to the second memory unit 107 and the controller 109.

The second memory unit 107 includes a selector 106
, And supplies it to the selector 106 via the line L102 as pre-match information. When the error comparison signal from the second comparing unit 108 is input to the second memory unit 107, the matching information stored in the second memory unit 107 is selected as the selection information.

Thus, as described above, the output of the perturbation checking unit 103 is stored in the first memory unit 105 and the second memory unit 107, or is invalidated by the result of the comparison by the first comparing unit 104. More specifically, when the error value from the perturbation check unit 103 is smaller than the previous error value on the line L101, the output of the perturbation check unit 103 is stored in the first memory unit 105 and the second memory unit 107,
Otherwise, the previous data on the lines L101 and L102 are stored in the first memory unit 105 and the second memory unit 107, respectively.
Is stored again.

The controller 109 is connected to the lines L10, L10
A plurality of control signals are generated based on the current contour information and the previous contour information input via the control unit 70. That is, the fact that all of the front contours stored in the contour storage unit 120 and sequentially output via the line L70 have been processed is determined by the mode selection unit 100 by the number of front contours sent via the line L70. , A first control signal S1 and a third control signal are generated. In response to the first control signal S1, the contour detection unit 160 generates new current contour information on the line L10. And the third
The control signal S3 is a signal for supplying the error value stored in the first memory unit 105 to the second memory unit 108. Further, the second control signal S2 is generated after the front contour information input via the line L70 is processed by the contour length comparing unit 101 and the centroid matching unit 102. Is a signal that causes the contour storage unit 120 to generate the next preceding contour information on the line L70 in response to When the error comparison signal is input from the second memory unit 108, the first mode selection signal M
S1 is provided, otherwise a second mode select signal MS2 is provided on line L20.

As described above, according to the example of FIG. 2, the mode selection unit 100 sequentially transmits the data from the contour storage unit 120 via the line L70 in accordance with the second control signal S2 generated from the controller 109. When selection processing is performed on all of the selected front contours, any one of the front contours stored in the contour storage unit 120 corresponding to the selection information becomes the optimum front contour for the current contour. Is selected as The selection information of the optimal front contour line is the line L
The signal is input to the motion compensator 110 through 30.

Referring back to FIG. 1, the motion compensator 110
Is based on the selection information input from the line L30,
Generate a predicted current contour. More specifically, the predicted current contour is generated by shifting the optimum front contour selected from the front contours stored in the contour storage unit 120 by the optimum displacement, and is generated via the line L40. The signals are supplied to the encoding unit 140 and the inter-reconstruction unit 190A, respectively.

The inter-encoding unit 140 inter-encodes the current contour input through the line L10 based on the selection information on the line L30 and the predicted current contour on the line L40. FIG. 3 is a detailed block diagram of the inter encoder 140 according to the preferred embodiment of the present invention.

The operation of the inter encoder 140 will be described below with reference to FIG. This inter encoding unit 140
Is composed of a matching unit 141, a deviation detecting unit 142, a video signal encoding unit 143, and a multiplexer (MUX) 144.

The matching section 141 superimposes the current contour and the predicted current contour as shown in FIG. 5, and supplies the overlapping contour to the deviation detecting section 142.

According to the preferred embodiment of the present invention, the deviation detector 142 calculates a difference between the current contour and the predicted current contour based on the overlapping contour.

FIG. 5 is a schematic diagram for explaining a process of detecting a deviation (ie, a single closed loop, for example, a convex portion) between the current contour and the predicted current contour. The deviation detecting unit 142 first draws a set of M radially extending first line segments that form an equiangular shape starting from a predetermined reference line from the common centroid T, and calculates the angle between two adjacent first line segments. With (P-1) second line segments (e.g., k1
~ K15). Here, the angular magnitude between two adjacent first line segments is 2π / M, where M is a positive integer greater than one. Thereafter, the deviation detecting unit 142 determines the intersection points of the M × P total line segments and the overlapping contour lines (for example, as shown in FIG. 5, A to P on the current contour line 10 and on the predicted current contour line 20). A ′ to P ′) are detected. Once all the intersections of the M × P line segments are detected, the deviation detection unit 142, for example, starts from the reference line and moves clockwise in the clockwise direction with respect to all the line segments and the current contour lines (eg, A, B, C, The error at each intersection with P) is determined. This error is calculated from the distance between the centroid T and the intersection of each line segment with the predicted current contour line, and the distance between the centroid T and the intersection point of each line segment with the current contour line (eg, TA). Is subtracted.
After determining the error values at all intersections on the current contour line 10 using the above-described deviation detection process, the error values are grouped into sets of P error values. The set of each error value is as follows.

[0039]

[Number 1] _{^{D1 = [d 1 1, d}} 1 2, ..., d 1 j, ..., d 1 p] D2 = [d 2 1, d 2 2, ..., d 2 j, ..., d 2 p] _{^{: DI = [d I 1,}} d I 2, ..., d I j, ..., d I p]: DM = [d M 1, d M 2, ..., d M j, ..., d M p]

Here, DI: the I-th array d ^I ₁ : an error value for the I-th first line segment d ^I _j : between the I-th first line segment and the (I + 1) -th first line segment Error value for the j-th second line segment d ^M _j : error value for the j-th second line segment between the M-th first line segment and the first first line segment (reference line segment) I: an integer from 1 to M j: an integer from 2 to P

According to another preferred embodiment of the invention, the M first segments are adaptively adjusted to the size of the overlapping contour.
That is, based on the distance between the centroid T and the intersection on the predicted current contour line 20, an average value is calculated for a predetermined M, and then the average value increases with respect to an appropriate threshold. Is adjusted so that M increases.

After that, the deviation detecting section 142 supplies the deviation information representing the set of arrays determined from the above process to the video signal encoding section 143.

The video signal encoding unit 143 includes, for example, 1
Using one-dimensional DCT (Discrete Cosine Transform) and one of the well-known quantization methods, each array included in the deviation information from the deviation detection unit 142 is converted into a set of quantized conversion coefficients, and MUX 144.

In the MUX 144, a set of quantized transform coefficients
The selection information input from the mode selection unit 100 via the line L30 and its stored data are output on the line L60 as inter-coded outline information.

As described above, the operation of the inter encoder 140 depends on the data received via the lines L30 and L40. Therefore, the selection information from the mode selection unit 100 via the line L30 or the line L40
If the predicted current contour from the motion compensation unit 110 is not input to the inter encoding unit 140 via the
There will be no output from 0.

Referring again to FIG. 1, the selector 150 and the switch 170 convert the intra-coded contour information or the inter-coded contour information according to the mode selection signal input via the line L20. Performs the function of transmitting. When the first mode selection signal MS1 is input,
Inter-encoded contour information is selected, sent to a transmitter (not shown) for transmission, and
0 is also supplied to the inter-decoding unit 180A. On the other hand, when the second mode selection signal MS2 is inputted, the intra-encoded contour information is selected and sent to the transmitter for transmission, and is also transmitted to the intra-decoding unit 180B via the switch 170. Supplied.

The inter-decoding unit 180A decodes the input inter-coded contour information using, for example, inverse quantization and inverse discrete cosine transform, and decodes the decoded inter-coded contour information. Inter reconstruction unit 1 as line information
90A.

Similarly, the intra decoding unit 180B decodes the input intra-encoded contour information,
It is supplied to the intra reconstruction unit 190B as decoded intra-coded contour information.

The inter-reconstructing section 190A reconstructs the contour based on the predicted current contour information input via the line L40 and the decoded inter-coded contour information from the inter-decoding section 180A. Generate information. The reconstructed contour information includes reconstructed contour video data and index data thereof, and the contour storage unit 1
20 is stored as new front contour information. The new front contour information is stored in the contour storage unit 12.
If other front contour information is already stored before being stored in 0, it is stored together therewith.

The intra reconstruction unit 190B generates reconstructed contour information based on the decoded intra-coded contour information from the intra decoding unit 180B.
As described above, it is stored in the contour storage unit 120 as new front contour information.

Thus, since the inter reconstruction unit 190A and the intra reconstruction unit 190B are selectively enabled by the input signals, the contour storage unit 120 stores the inter reconstruction unit 190A or the intra reconstruction unit 190A.
The output from B will be stored selectively.

When all the outline video data included in the video frame signal is once encoded by the above-described outline encoding process, a new video frame signal is input to the outline detection unit 160, and as described above. Then, the encoding process continues.

The above-described contour coding method can also be used for a contour coding process between two video frame signals, that is, a current frame signal and a previous frame signal.

Although the preferred embodiments of the present invention have been described above, it is obvious that various modifications can be made without departing from the scope of the present invention.

[0055]

Therefore, according to the present invention, the contour of an object represented by a video signal is adaptively adjusted using a contour motion estimation technique based on the difference between the previous contour and the current contour. The amount of data to be encoded and transmitted can be further reduced.

[Brief description of the drawings]

FIG. 1 is a schematic block diagram of a contour coding apparatus according to the present invention.

FIG. 2 is a detailed block diagram of a mode selection unit shown in FIG.

FIG. 3 is a detailed block diagram of an inter encoder shown in FIG.

FIG. 4 is a schematic diagram for explaining a process of detecting an error value of an optimum contour according to the present invention, which is composed of A and B.

FIG. 5 is a schematic diagram for explaining an inter encoding process performed by an inter encoding unit.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Current contour 20 Predicted current contour 30 Optimal current contour 100 Mode selection unit 101 Contour length comparison unit 102 Centroid matching unit 103 Perturbation check unit 104 First comparison unit 105 First memory unit 106 Selector 107 Second memory Unit 108 second comparing unit 109 controller 110 motion compensating unit 120 contour storing unit 130 intra encoding unit 140 inter encoding unit 141 matching unit 142 deviation detecting unit 143 video signal encoding unit 144 MUX 150 selecting unit 160 contour detecting unit 170 Switch 180A Inter decoding unit 180B Intra decoding unit 190A Inter reconstructing unit 190B Intra reconstructing unit AP, C1 to C12, P1 to P12 Intersection T Centroid

Claims (17)

[Claims] An adaptive contour coding method for adaptively coding a video frame signal having a plurality of contours represented by contour video data representing the position of a contour pixel in contour units. A first step of allocating index data to each of the contours in the processing order, a second step of generating contour image data representing the contour video data of the current contour and index data, A third step of generating front contour information representing contour image data and index data of each front contour processed before the processing of the contour, and the current contour information and all the front contours for the front contour. Based on the front contour information, the selection information and the first or second
A fourth step of generating a mode selection signal, a fifth step of inter-coding the current contour information based on the selection information in response to the first mode selection signal, and a second step of responding to the second mode selection signal A sixth step of intra-encoding the current outline information, and a seventh step of generating the inter-encoded outline information or the intra-encoded outline information as encoded outline information. And an adaptive contour coding method. 2. The method according to claim 1, wherein the fourth step is to assign one of the front contour lines as a target front contour line, and the length of the current contour line and the length of the target front contour line. And calculating a candidate displacement representing a spatial displacement between the current contour line and the target front contour line in accordance with the comparison signal. Fourth to calculate
Step 4-4 of shifting the entire contour pixel on the target front contour by the candidate displacement to generate a predicted contour for the target front contour, and along the current contour. Shifting the predicted contour in the horizontal and vertical directions by one or more pixels to detect an optimal contour that provides an error value that minimizes the difference between the predicted contour and the current contour. Generating an error value for the optimal contour line and an optimal displacement obtained by adding the candidate displacement to a spatial displacement between the optimal contour line and the current contour line; Step 4-6 of repeating the above process for all remaining front contours, and comparing the error values of the optimum contour with respect to the entire front contour to each other, Any one of
A fourth-fourth step which is generated as an optimal front contour line having a minimum error value; and comparing the error value with respect to the optimal front contour line with a predetermined threshold value, and based on the comparison, the first or second step.
A step 4-8 of generating a mode selection signal; and a step 4-9 of generating the optimum displacement and the index data of the optimum front contour line based on the first mode selection signal. The adaptive contour coding method according to claim 1, wherein: 3. The comparison signal is generated when the difference between the length of the current contour and the length of the previous contour is within a predetermined range of the length of the current contour. 3. The adaptive contour coding method according to claim 2, wherein: 4. The method according to claim 1, wherein the error value of the optimal front contour is equal to or less than the predetermined threshold.
4. The adaptive contour coding method according to claim 3, wherein a mode selection signal is generated, and if not, the second mode selection signal is generated. 5. The adaptive contour according to claim 2, wherein the error value of the optimal contour is defined as a region of a non-overlapping portion between the current contour and the optimal contour. Line encoding method. 6. The method according to claim 2, wherein the error value of the optimal contour is defined as the number of pixels existing in a non-overlapping area between the current contour and the optimal contour. Adaptive contour coding method. 7. In order to determine the error value of the optimal contour, the step 4-5 is performed in advance from a centroid determined by averaging position coordinates of all the contour pixels on the current contour. A 4-5-1 process of drawing a set of K radial line segments (K is a positive integer) forming an equal angle starting from the reference line, and the K radial line segments, the optimal contour line, A 4-5-2 step of detecting an intersection point where the current contour line intersects; and a 4-5-3 step of calculating, as the error value, a sum of absolute values of displacements between the intersection points on each radial line segment. 3. The adaptive contour coding method according to claim 2, further comprising the steps of: 8. An adaptive contour encoding apparatus for adaptively encoding a video frame signal having a plurality of contours represented by contour video data representing the position of a contour in contour units. Indexing each of said contours in their processing order;
Index assigning means for assigning index data to the contour, first contour information generating means for generating contour image data representing the contour image data of the current contour and index data, and Second contour information generating means for sequentially storing and generating contour image data and index data representing the contour data of each front contour processed before the processing, and the current contour information and all the previous contour information. A mode selection means for generating selection information and a mode selection signal based on the previous contour information on the contour, and inter-coding the current contour information in accordance with the selection information to obtain an inter-coded contour Inter-coding means for generating line information, intra-coding the current contour information, and intra-coding means for generating intra-coded contour information; Selecting means for selecting the inter-encoded contour information or the intra-encoded contour information as encoded contour information to be transmitted according to the mode selection signal. Adaptive contour coding device. 9. The mode selection means calculates a length of the current contour and a length of the previous contour,
First comparing means for comparing and generating a comparison signal; and for each of the front contour lines according to the comparison signal,
Candidate displacement calculating means for calculating a candidate displacement representing a spatial displacement between the current contour and the front contour, and shifting the entire contour pixel on the front contour by the candidate displacement, Predicted contour generating means for generating a predicted contour for each of the front contours; and shifting the predicted contour in the horizontal and vertical directions by one or more pixels along the current contour, Detecting, for the predicted contour line corresponding to each of the lines, an optimal contour line that provides an error value that minimizes a difference between the current contour line and the predicted contour line; An optimal contour detecting means for generating an optimal displacement obtained by adding the candidate displacement to a spatial displacement between the optimal contour and the current contour; and an optimal contour corresponding to each of the front contours. Compare contour error values with each other An optimal front contour detecting means for detecting any one of the optimal contours as an optimal front contour minimizing an error value; and determining the error value for the optimal front contour in advance. Second comparing means for comparing the threshold value with the threshold value and generating the mode selection signal based on the threshold value; and generating the optimum displacement and the index data of the optimum front contour line as the selection information according to the mode selection signal. 9. The adaptive contour coding apparatus according to claim 8, further comprising selection information generating means. 10. The comparison signal is generated when the difference between the length of the current contour and the length of the previous contour is within a predetermined range of the length of the current contour. 10. The adaptive contour coding apparatus according to claim 9, wherein: 11. The adaptive contour according to claim 9, wherein the mode selection signal is generated when the error value of the optimal front contour is equal to or less than the predetermined threshold. Encoding device. 12. The error value of the optimal contour is
The adaptive contour coding apparatus according to claim 9, wherein the adaptive contour coding apparatus is defined as a region of a non-overlapping portion between the current contour and the optimal contour. 13. The error value of the optimal contour is
10. The adaptive contour encoding apparatus according to claim 9, wherein the number of pixels is defined as a number of pixels existing in a non-overlapping area between the current contour and the optimal contour. 14. The method according to claim 1, wherein the optimal contour detecting means determines the error value of the optimal contour from a centroid obtained by averaging position coordinates of all the contour pixels on the current contour. Drawing means for drawing a set of K radial line segments (K is a positive integer) forming an equal angle starting from a predetermined reference line for the first time from a predetermined reference line, and the K radial line segments and the optimal Intersection detection means for detecting an intersection of a contour line and the current contour line, and error value calculation means for calculating the sum of absolute values of displacement between intersection points on the radial line segment as the error value. 10. The adaptive contour coding device according to claim 9, wherein: 15. The inter-encoding unit predicts a front contour having the same data as the index data of the selection information based on the front contour information and the selection information for the entire front contour. A predicted current contour generating means that is selected and generated as a current contour, an overlapping means for generating the overlapping contour by superimposing the predicted current contour and the current contour, and Deviation information detection means for detecting deviation information representing a shape difference between the predicted current contour and the current contour, conversion means for converting the deviation information, the converted deviation information and the selection information And a contour line information generating means for generating the contour line information as the inter-encoded contour line information. 16. The deviation information detecting means, wherein: a drawing means for drawing a plurality of radial line segments from a centroid of the current contour; and a radial drawing means for each of the current contour and the predicted current contour. Intersection detection means for detecting an intersection between each line segment and the two contour lines; and calculating an error at the intersection of the current contour line based on a difference between the two intersection points on each of the line segments, and calculating the deviation information. 16. The adaptive contour coding apparatus according to claim 15, further comprising: an error calculating means for generating the following. 17. The distance between the centroid of the current contour and the intersection of the current contour is determined from the distance between the centroid of the current contour and the intersection of the predicted current contour. 17. The value obtained by subtracting
5. The adaptive contour coding device according to claim 1.